AI Code of Conduct: Comment and Analysis

Some thoughts from Rory Yates at EIS, after the announcement of a new insurance industry Code of Conduct, regarding AI usage in insurance products;

This code is a set of relatively simple checks and balances we can all make, or take responsibility for, as we apply AI to insurance.

Given my background and experience in launching frameworks and standards for responsibility, several delegates at the launch asked “will the code be successful?’. I answered simply,”Yes.” Here’s why.

We are living in a time of infinite possibilities, where new technologies are emerging constantly. However, while nothing compares to the power of AI, there are distinct parallels between all technological advances. In that, only a few people really get the new tech at first, and yet many others try to apply it, and get the value they believe is possible.

At this time, acting responsibly is critical. Not just because it’s something most want, but because these codes or standards help us plot a successful path through the development process. Also, the reality is, we all know regulation is coming and will catch up quickly. And there are already examples of US insurers Humana, Cigna, and UnitedHealthcare who are facing class actions from consumers and their estates for allegedly deploying advanced technology to deny claims.

So, it’s in everyone’s interest that when asked why a decision has been taken, the answer isn’t because the machine says so. It’s because ‌accountability sits with humans. Which is why this code aims to make sure that insurers are fair, transparent, and logical when using AI in claims. As well as guaranteeing that customers have the means to challenge decisions made by machines.

It’s important to note that this isn’t the only one. There are many codes of conduct emerging for AI, e.g. The International Code of Conduct for Organizations Developing Advanced AI Systems aims to promote safe, secure, and trustworthy AI worldwide. It’ll provide voluntary guidance for actions by organizations developing the most advanced AI, including the most advanced foundation models and generative AI systems.

The UK government has launched its own code of conduct for AI systems used by the NHS. This one is specifically about insurance claims, which is important because generalised codes can often be too broad. At this stage in AI’s journey, we need to work hard to demystify it and help people who’ll be held to account to understand what they’re exposing themselves to. Both in terms of risk and reward, and how to eliminate the former and promote the latter.

I was honored to be part of the development process, and I have ‌deep gratitude to Eddie Longworth and the team for pulling the working group and code together. There’s no reason not to sign up if you’re in insurance. This isn’t asking anyone to do anything you, as a consumer, wouldn’t expect a responsible company to be doing, and I’m extremely grateful to be in an industry that responds in this way.

The immediate uptake and support has been spectacular. Long may it continue.